Content-based image indexing and retrieval : feature extraction and feature similarity metrics

Content-based image retrieval is emerging as a significant research area with application to digital libraries and multimedia databases. This thesis presents the development of a content-based image indexing and retrieval system and methodologies studied and investigated through the development. This the-sis focuses on two areas: feature extraction and feature similarity metrics. Feature extraction is based on multiresolution analysis of images. Feature similarity is based on several new similarity metrics developed or adopted in this work, namely the two-level similarity metric, Nearest Feature Line distance and its extension.Master of Engineerin